[clang] [llvm] [IR] Allow fast math flags on fptrunc and fpext (PR #115894)

[Joshua Cranmer via llvm-commits]

================
@@ -3730,10 +3730,10 @@ Fast-Math Flags
 
 LLVM IR floating-point operations (:ref:`fneg <i_fneg>`, :ref:`fadd <i_fadd>`,
 :ref:`fsub <i_fsub>`, :ref:`fmul <i_fmul>`, :ref:`fdiv <i_fdiv>`,
-:ref:`frem <i_frem>`, :ref:`fcmp <i_fcmp>`), and :ref:`phi <i_phi>`,
-:ref:`select <i_select>`, or :ref:`call <i_call>` instructions that return
-floating-point types may use the following flags to enable otherwise unsafe
-floating-point transformations.
+:ref:`frem <i_frem>`, :ref:`fcmp <i_fcmp>`, :ref:`fptrunc <i_fptrunc>`,
+:ref:`fpext <i_fpext>`), and :ref:`phi <i_phi>`, :ref:`select <i_select>`, or
+:ref:`call <i_call>` instructions that return floating-point types may use the
----------------
jcranmer-intel wrote:

Agree, `fptrunc` and `fpext` need updating to at least show where the fast-math flags may occur.

https://github.com/llvm/llvm-project/pull/115894